journal-of-management-sciences-china

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates as a purely instructional template for academic guidance. It does not include executable code, scripts, or external dependencies.
  • [DATA_EXPOSURE]: The skill references internal resource files located at ../../resources/source-basis.md and ../../resources/official-source-map.md. These are standard relative paths used within the agent's environment for referencing verified source mappings and do not constitute sensitive data exposure.
  • [PROMPT_INJECTION]: No attempts to override system prompts, bypass safety filters, or extract system instructions were detected. The language is consistently professional and academic.
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to retrieve and verify submission guidelines from official journal websites. While this involves processing external data, the skill's capabilities are limited to generating structured text output, presenting a negligible risk of instruction injection via external content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 09:33 AM
Security Audit — agent-trust-hub — journal-of-management-sciences-china